Mammographic screening leads to many women being diagnosed with ductal carcinoma in situ [DCIS], yet we cannot accurately predict which lesions will undergo malignant progression to invasive ductal carcinomas [IDC] or effectively block this transition. Studies of human breast biopsies have implicated in this process cysteine cathepsins V/L2 and B in tumor cells and macrophages and cathepsins F, K and L in myoepithelial cells/[myo]fibroblasts. Aberrant signal transduction, for example through p21-activated kinase 1 [PAK1], may contribute to increased pericellular proteolysis. Our working hypothesis is that the transition from pre-invasive DCIS to invasive carcinomas and the rapid progression of some DCIS lesions are mediated through alterations in proteolytic pathways in DCIS cells and DCIS-associated cells, and that dysregulated PAK1 contributes to the induction of these aberrant proteolytic pathways. To test this hypothesis, we will recapitulate the transition from pre-invasive DCIS to invasive carcinoma using in vitro and in vivo progression models that we have designated MAME for mammary architecture microenvironment engineering. In these models, we will use isogenic MCF10 cell lines [AT1, DCIS1 and CA1d] and two human DCIS cell lines [SUM-102 and SUM-225]. Our specific aims are to: 1. Modulate expression and activity of cysteine cathepsin V or B in the isogenic and SUM DCIS cell lines, both by direct targeting and through intervention at the level of PAK1, and determine using the in vitro MAME model whether the invasive phenotype is altered; 2. Determine using the in vitro MAME model whether the invasive phenotype can be altered by co-culturing modified cells from Aim 1 with myoepithelial cells, [myo]fibroblasts or both cell types, using wild-type cells and ones in which expression and activity of cysteine cathepsin F, K or L have been modulated; 3. Determine using the in vivo MAME model whether the malignant phenotype of xenografts of modified cells from Aim 1 can be altered by simultaneous implantation of myoepithelial cells, [myo]fibroblasts or both cell types, using wild-type cells and ones in which expression and activity of cysteine cathepsin F, K or L have been modulated; and 4. Screen [via our Hu/Mu ProtIn chip] the in vivo MAME model for proteolytic pathways that may contribute to the transition from DCIS to IDC and use the in vitro MAME model to define functional changes with libraries of reagents from the Center on Proteolytic Pathways. Validating, in the context both of the tumor and its microenvironment, proteases key to progression of DCIS to IDC, and kinase pathways that regulate them, should identify potential targets for therapeutic intervention as well as biomarkers to distinguish DCIS lesions that will rapidly progress to IDC. PUBLIC HEALTH RELEVANCE: Proteases and kinases are the subject of intensive efforts by the pharmaceutical industry to develop new treatment strategies for human diseases, including cancer. Our proposed studies will discover and validate protease pathways that are active in the tumor microenvironment and that mediate the transition to a full-blown malignancy, and kinase pathways that regulate these protease pathways. We anticipate that our studies will identify biomarkers to distinguish premalignant lesions that will progress to invasive cancers and define targets that will abrogate that progression.